This application requests 2 years of funding to support John Medaglia's pre-doctoral training in functional neuroimaging data modeling and clinical neuroscience research. The proposed research will apply novel techniques to understand the role of the cerebellum as a latent support mechanism for working memory performance following moderate-to-severe traumatic brain injury (TBI) to better understand the processes underlying cognitive deficits and recovery. This project is distinct from traditional fMR research that attempts to isolate regional differences between individuals with TBI and matched healthy controls in that it affords explicit quantitative and qualitative examinations of how neura networks are affected by injuries. This proposal consists of 3 aims, each with an associated experimental approach. Specific Aim 1 is to examine the role of a traditionally understudied region, the cerebellum, in a distributed working memory (WM) system with a critical role in learned timing, pattern detection, associative learning, and speed of information processing. It is hypothesized that the cerebellum will be highly related to previously identified regions involved in WM (i.e., the dorsolateral prefrontal cortex, anterior cingulate cortex, and parietal cortex) during task performance and that the strengths of these relationships will predict performance, particularly those between the cerebellum and the prefrontal cortex. Specific Aim 2 is to test the hypothesis that the primary large-scale networks observed during WM tasks (i.e., involving the dorsolateral prefrontal cortex, anterior cingulate cortex, parietal cortex, and cerebellum) in controls will be disrupted in TBI and that disruption will predict behavioral performance. Importantly, this extends beyond Aim 1 by considering the joint functions of large networks as important to behavior as opposed to each part in isolation. It is hypothesized that controls will have more closely interrelated functional networks loosely constrained by anatomical connections, whereas individuals with TBI will have fractionated networks with specific disruptions in cerebellar and prefrontal functional connections that are predictive of cognitive dysfunction. Aim 3 will seek to corroborate functional findings in brain structural connectivity using diffusion tensor imaging. It is hypothesized that anatomical integrity will predict the degre of functional connectivity across the brain as well as specific functional relationships between the dorsolateral prefrontal cortex and parietal cortex, which have anatomical connections with the cerebellum. The results from this proposal will advance our understanding of the mechanisms of how the brain responds to injury as a neurocognitive system as opposed to previous findings that do not account for the complex relationships among regions in the brain during cognitive processing. This is a critical step toward future aggressive treatment of severe injury because it will aid our understanding of how disrupted activity in certain parts of the neurl system affects others, which may have critical implications for neurosurgery, medication, and cognitive rehabilitation. This proposal will also prepare the Applicant with advanced expertise in signal analysis, linear and nonlinear equation modeling, the utility of graph theory in understanding the brain, and structural connectivity techniques which will provide the basis for a productive independent research career.